Novel associations between MTDH gene polymorphisms and invasive ductal breast cancer: a case–control study

TL;DR

This study finds that certain genetic variations in the MTDH gene are linked to a higher risk of invasive ductal breast cancer.

Contribution

The study identifies novel associations between specific MTDH gene polymorphisms and increased risk of invasive ductal breast cancer.

Findings

Three MTDH SNPs (rs1311, rs16896059, rs2449512) are significantly associated with increased risk of invasive ductal breast cancer.

Certain genotypes of these SNPs are linked to higher IDC risk in specific patient subgroups based on age, menopausal status, and tumor characteristics.

Haplotypes TAA, TAG, and TGG are significantly associated with increased IDC risk compared to the reference haplotype TGA.

Abstract

To reveal the contributing effects of MTDH gene SNPs in the risk of invasive ductal breast cancer (IDC).

A case–control study was conducted, recruiting a total of 300 cases of IDC and 565 cancer-free controls from East China. Genotyping of three single-nucleotide polymorphisms (SNPs) in the MTDH gene was performed. Genomic DNA was extracted from peripheral blood samples of patients. The three SNPs (rs1311 T > C, rs16896059 G > A, rs2449512 A > G) in the MTDH gene were selected for detection using a TaqMan real-time polymerase chain reaction assay. The association between MTDH and the risk of IDC was analyzed employing an epidemiology case–control study and a multinomial logistic regression model.

Among the three evaluated SNPs, rs1311 T > C, rs16896059 G > A, and rs2449512 A > G demonstrated a significant association with an increased risk of IDC. Furthermore, stratified analysis revealed that individuals carrying the rs1311 CC genotype, rs16896059 GA/AA genotypes, and rs2449512 GG genotype were more susceptible to developing IDC in subgroups of patients younger than 53 years, without family history of IDC, pre-menopause status, clinical stage 2, high grade, with no distant metastasis or invasion, Her2-positive type, ER positive, PR positive, and Ki67 cells less than 10%. However, carriers of the rs16896059 GA/AA genotypes and rs2449512 GG genotype had an elevate the risk of IDC in patients with tumor size larger than 2 cm, post-menopause status, clinical stage 3, with invasion, lymph node infiltration, ER negative, PR negative, Her2 negative, and Ki67 cells exceeding 10%. Compared to the reference haplotype TGA, haplotypes TAA, TAG, and TGG were significantly associated with an increased IDC risk.

In this study, we demonstrated a significant association between MTDH gene polymorphisms and an increased risk of IDC. Moreover, our findings suggested that MTDH gene polymorphisms could serve as a potential biomarker for IDC subtyping and therapeutic selection.

Full text

Introduction

Recent research indicates that breast cancer has emerged as the most prevalent malignant tumor, ranking second only to lung cancer as the leading cause of cancer-related death among women worldwide [1]. The incidence of breast cancer continues to escalate annually, although the prevalence varies significantly among countries due to differences in age and lifestyle factors [2]. Breast cancer is histopathologically classified into several subtypes, with invasive ductal carcinoma (IDC) being the most common subtype accounting for approximately 70–80% of cases. Based on the expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor (Her2), breast cancer can be categorized into four molecular subgroups: luminal A type, luminal B type, Her2-positive type, and basal-like type [3]. Various factors such as family history, breastfeeding, obesity, and environmental influences, contribute to the risk of developing breast cancer. Additionally, genetic variations play a crucial role in both the initiation and progression of breast cancer [4]. Single nucleotide polymorphism (SNP) represents one form of genetic variation that has been extensively studied regarding its association with susceptibility to breast cancer development [5].

SNPs in metastasis-related genes function as vital roles in the carcinogenesis of breast cancer, as demonstrated by various studies. The association between the variations of these genes and breast cancer has been corroborated by research, such as the identification of the polymorphism of tissue inhibitor of metalloproteinase-2 (TIMP-2) as a risk factor for breast cancer [6]. Li Z, et al. also reported a link between the vascular endothelial growth factor (VEGF) gene − 634G/C polymorphism and an increased risk of breast cancer [7]. Moreover, SNPs in ERBB3 and BARD1 genes were found to indicate a poorer prognosis for HER2-positive breast cancer patients [8]. The NF-κB1 rs28362491 polymorphism was associated with an increased risk of breast cancer in Lower Northern Thailand [9], while a positive association between NF-κB rs3774937 and breast cancer was observed in the Middle Eastern-North African population [10]. Alanazi MS, et al. investigated the association of Wnt signaling pathway gene polymorphisms with breast cancer and found that the SNP in beta-catenin was positively related to the risk of breast cancer in Saudi patients [11]. Collectively, these findings corroborate the notion that polymorphisms of metastasis-related genes are associated with the risk of breast cancer.

Metadherin (MTDH), also known as astrocyte elevated gene 1 (AEG-1), is located at chr8q22.1 and functions as a metastatic adhesion protein, making it a therapeutic target in various cancers [12]. MTDH is overexpressed in multiple cancers, including breast cancer [13]. A close relationship has been observed between MTDH expression and the poor prognosis of breast cancer patients [14]. The upregulation of MTDH is associated with a better prognosis of Her2-positive breast cancer patients [15]. MTDH promotes metastasis and invasion by interacting with VEGF [16], Twist1 [17], NF-κB [18], and other genes involved in metastasis-related signaling pathways. Inhibition of MTDH reduces paclitaxel resistance in breast cancer cells [19]. Moreover, suppressing MTDH activity hinders the metastatic potential of breast cancer cells [20]. To date, only one study has revealed a negative association between MTDH (− 470G > A) polymorphism and ovarian cancer susceptibility [21]. Given that numerous metastasis-related gene polymorphisms are known to be associated with an increased risk of breast cancer, the impact of MTDH polymorphisms on breast cancer susceptibility has not yet been reported.

In the current investigation, a total of three SNPs were selected to evaluate the association between MTDH polymorphisms and IDC. The study was designed as a case–control analysis, utilizing samples from Eastern China.

Materials and methods

Study subjects and data collection

A sample of 300 breast cancer patients and 565 age-matched and ethnicity-matched healthy controls from Eastern China, aged 24 to 96 years old, median age was 53, was recruited from The First People’s Hospital of Linping District, Hangzhou City, Zhejiang Province. This study collected breast cancer cases from January 2013 to May 2020. The required sample size was estimated based on a similar study [22]. Cases were selected according to pathological diagnosis, while controls were recruited from health adult women undergoing physical examination at The First People’s Hospital of Linping District, Zhejiang Province. Exclusion criteria included women with other malignancies, gynecological diseases, endocrine system disorders, or who were breastfeeding.

The comprehensive clinical and biological features of breast cancer patients, including age, clinical stage, tumor size, pathological grade, family history, molecular type, lymph node involvement, invasion, and metastasis were comprehensively analyzed and tabulated in Table 1. Table 1. Characteristics of the participantsCases (300)Controls (565)pAge (y)53.6 ± 11.752.3 ± 12.60.126BMI (kg/m2)23.1 ± 4.822.8 ± 4.20.246Family history0.001 Yes18 (6.0%)10 (1.8%) No282 (94.0%)555 (98.2)Menopausal age (y)11.6 ± 2.411.9 ± 1.80.425Pausimenia0.236 Post-menopause178 (59.3%)248 (40.3%) Pre-menopause132 (40.7%)217 (59.7%)Tumor size  < 2 cm174 (58.0%)  ≥ 2 cm126 (42.0%)Clinical stage 124 (8.0%) 2183 (61.0%) 393 (31.0%)Pathological grade Low290 (96.7%) High10 (3.3%)Distance metastasis Yes11 (3.6%) No289 (96.4%)Invasion Yes238 (79.3%) No62 (21.7%)Node infiltration Yes102 (34.0%) No198 (67.0%)Molecular type LuminalA123 (41.0%) LuminalB82 (27.3%) Her2-positive47 (15.7%)B asal-like48 (16.0%)ER expression Negative93 (31.0%) Positive207 (69.0%)PR expression Negative95 (31.7%) Positive205 (68.3%)Her2 expression Negative172 (57.3%) Positive128 (42.7%)Ki67 expression  < 10%219 (73.0%)  ≥ 10%81 (27.0%)

The study was granted ethical approval by the Ethics Committee of The First People’s Hospital of Linping District, Zhejiang Province (reference number: 2018-152), and written informed consent was obtained from all samples enrolled in accordance with the Declaration of Helsinki.

MTDH SNPs selection and genotyping

The NCBI dbSNP database (http://www.ncbi.nlm.nih.gov/projects/SNP) and SNPinfo (http://snpinfo.niehs.nih.gov/snpfunc.htm) online software were utilized to identify selected potentially functional SNPs. The selection criteria were based on published studies, as follows [23, 24]: the minor allele frequency (MAF) of SNPs identified in HapMap no less than 5% for Chinese Han subjects; SNPs were located in the 5ʹ and 3ʹ untranslated region, exons and the junctions of exon and intron of the MTDH gene, as well as in regions with low linkage disequilibrium (R2 < 0.8). Three SNPs (rs1311 T > C, rs16896059 G > A, rs2449512 A > G) in the MTDH gene were selected. Rs1311 and rs2449512 are situated in the 3ʹUTR of MTDH, potentially serving as binding sites for miRNAs. rs16896059, on the other hand, is located in the promoter of MTDH and is predicted to be a binding site for transcriptional factors. A quantity of 1 μg genomic DNA was extracted from IDC patients’ and control samples’ peripheral blood. The reaction system and conditions of the Taqman real-time PCR assay were in accordance with a published reference [25]. Taqman probes of rs1311 (Assay ID; C_11331064_30), rs16896059 (Assay ID: C_27850029_10), and rs2449512 (Assay ID: C_15799756_10) were purchased from Thermo Fisher. To ensure the accuracy of the genotyping results, 10% of the samples were randomly selected for genotyping by DNA sequencing. A concordance rate of 100% was achieved for the quality control samples.

Statistical analysis

The goodness-of-fit χ^2^ test was employed to assess whether the znalyzed SNPs in MTDH gene deviated from Hardy–Weinberg equilibrium (HWE) among controls. A two-sided χ^2^ test was conducted to compare demographic variables and genotype frequencies of patients and controls. Odds ratios (ORs), age-adjusted ORs, and their corresponding 95% confidence intervals (CIs) for the association between SNPs and susceptibility of breast cancer were counted using unconditional logistic regression analyses. Furthermore, a combination of rs1311 T > C, rs16896059 G > A, and rs2449512 A > G was considered a haplotype. Unphased genotype data were utilized to determine haplotype frequencies and individual haplotypes. Logistic regression analysis also assisted in calculating haplotype frequencies and distinct haplotypes, with adjustment for age. The haplotype with the highest frequency was used as the reference group to calculate ORs for haplotypes associated with IDC risk [26]. All statistical analyses were conducted through the SAS statistical package (version 9.1; SAS Institute, Cary, NC). All P values in this study were two-sided, and a P value < 0.05 was supposed to be statistically significant.

Results

Population characteristics

The summarized information of the demographic and clinical features of IDC patients and healthy controls was listed in Table 1. No significant differences were observed between Eastern Chinese women with IDC and the controls in terms of age (P = 0.1263), BMI (P = 0.2457), and menopausal age (P = 0.4251). However, a significant difference was noted between the premenopausal women who had IDC compared to the control group (P = 0.236). Among IDC cases, tumors less than 2 cm accounted for 58% (174 cases) while those greater than or equal to 2 cm accounted for 42% (126 cases). Regarding clinical stage, there were 24 cases (8%) at stage1183 cases (61%) at stage 2, and 93 cases (31%) at stage 3. In terms of pathological grade, high-grade tumors were present in three patients (3.3%), while low-grade tumors were present in 96.7% (290 cases). Eleven patients (3.6%) had distant metastasis, while 96.4% (289 cases) did not. Of the 208 IDC cases (79.3%) analyzed, 62 cases (21.7%) had no invasion. One hundred and two cases (34.0%) had node infiltration, while 198 cases (67.0%) did not. Regarding molecular subtype, 41.0% (123 cases) were luminal A type, 27.3% (82 cases) were luminal B type, 15.7% (47 cases) were Her2-positive type, and 16.0% (48 cases) were basal-like type. ER was expressed in 31.0% (207) and negatively expressed in 69.0% (93) cases. PR was expressed in 68.3% (205) and negatively expressed in 31.7% (95) cases. The expression of Her2 was positive in 42.7% (128 cases) and negative in 57.3% (172 cases). Lastly, 27.0% (81 cases) had more than 10% positive Ki67 expressed cells, while 73.0% (219 cases) had less than 10% positive Ki67 expressed cells in IDC tissues.

Association of MTDH gene polymorphisms with IDC risk

The genotype frequencies of MTDH associated with IDC risk were presented in Table 2. In the single-locus analysis, carriers of the rs1311 (CC vs. TT: adjusted OR = 2.775, 95% CI 1.114–6.910, P = 0.0283), rs16896059 (GA vs. GG: adjusted OR = 1.916, 95% CI 1.114–3.295, P = 0.0187; AA vs. GG: adjusted OR = 31.656, 95% CI 4.155–241.196, P = 0.0009), and rs2449512 (GG vs. AA: adjusted OR = 4.504, 95% CI 2.093–9.691, P = 0.0001) variant alleles were found to contribute to an elevated risk of IDC. Table 2. Logistic regression analysis of associations between MTDH polymorphisms and IDC susceptibilityGenotypeCasesControlsp^a^Crude ORpAdjusted ORp^b^(N = 300)(N = 565)(95% CI)(95% CI)^b^rs1311 T > C (HWE = 0.642) TT244 (81.33)448 (79.29)1.001.00 TC43 (14.33)109 (19.29)0.724 (0.492–1.065)0.10140.710 (0.479–1.052)0.0875 CC13 (4.33)8 (1.42)2.981 (1.219–7.290)0.01672.775 (1.114–6.910)0.0283 Additive691170.00781.040 (0.774–1.397)0.79381.013 (0.750–1.368)0.9315 Dominant56 (18.67)117 (20.71)0.47500.879 (0.616–1.253)0.47520.856 (0.596–1.229)0.3992 Recessive287 (95.67)557 (98.58)0.00803.151 (1.291–7.689)0.01172.944 (1.185–7.313)0.0200rs16896059 G > A (HWE = 0.403) GG254 (84.67)534 (94.51)1.001.00 GA29 (9.67)30 (5.31)2.032 (1.194–3.459)0.00901.916 (1.114–3.295)0.0187 AA17 (5.67)1 (0.18)35.733 (4.730–269.948)0.000531.656 (4.155–241.196)0.0009 Additive6332 < 0.00012.933 (1.961–4.387) < 0.00012.772 (1.839–4.178) < 0.0001 Dominant46 (15.33)31 (5.49) < 0.00013.120 (1.932–5.038) < 0.00012.902 (1.781–4.730) < 0.0001 Recessive283 (94.33)564 (99.82) < 0.000133.873 (4.486–255.775)0.000630.116 (3.953–229.426)0.0010rs2449512 A > G (HWE = 0.103) AA221 (73.67)449 (79.47)1.001.00 AG58 (19.33)105 (18.58)1.122 (0.784–1.607)0.52871.081 (0.749–1.560)0.6757 GG21 (7.00)11 (1.95)3.879 (1.838–8.187)0.00044.504 (2.093–9.691)0.0001 Additive1001270.00071.477 (1.136–1.921)0.00361.501 (1.148–1.964)0.0030 Dominant79 (26.33)116 (20.53)0.05191.384 (0.997–1.921)0.05241.374 (0.983–1.920)0.0633 Recessive279 (93.00)554 (98.05)0.00023.791 (1.802–7.973)0.00044.436 (2.068–9.515)0.0001^a^χ^2^ test for genotype distributions between breast cancer cases and controls^b^Adjusted for age

Stratification analysis of identified SNPs

The influence of the selected MTDH polymorphisms (rs1311 T > C, rs16896059 G > A, rs2449512 A > G) on specific subtypes of IDC was further examined (Table 3). For rs1311 T > C, a significantly increased risk effect was observed among patients aged younger than 53 years (adjusted OR = 7.997, 95% CI 1.527–41.880, P = 0.0139), with tumor size less than 2 cm (adjusted OR = 3.554, 95% CI 1.323–9.548, P = 0.0119), no family history (adjusted OR = 2.794, 95% CI 1.106–7.061, P = 0.0298), pre-menopause (adjusted OR = 4.609, 95% CI 1.587–13.386, P = 0.0050), clinical stage 2 (adjusted OR = 3.187, 95% CI 1.183–8.588, P = 0.0219), high pathological grade (adjusted OR = 1.052, 95% CI 1.228–7.586, P = 0.0163), no distance metastasis (adjusted OR = 3.058, 95% CI 1.233–7.584, P = 0.0159), no invasion (adjusted OR = 2.700, 95% CI 1.037–7.026, P = 0.0419), no node infiltration (adjusted OR = 2.837, 95% CI 1.065–7.588, P = 0.0370), luminal B type (adjusted OR = 4.268, 95% CI 1.312–13.887, P = 0.0159), Her2-positive type (adjusted OR = 4.402, 95% CI 1.118–17.339, P = 0.0341), ER positively expressed (adjusted OR = 2.803, 95% CI 1.043–7.530, P = 0.0410), PR positively expressed (adjusted OR = 2.831, 95% CI 1.053–7.609, P = 0.0391), Her2 positively expressed (adjusted OR = 4.402, 95% CI 11.590–12.193, P = 0.0043), and Ki67 expressed cells < 10% (adjusted OR = 2.923, 95% CI 1.117–7.653, P = 0.0289). Table 3. Stratification analysis of MTDH polymorphisms with IDC susceptibilityVariablesrs1311 (cases/controls)Adjusted OR^a^p^a^rs16896059 (cases/controls)Adjusted OR^a^p^a^rs2449512 (cases/controls)Adjusted OR ^a^p ^a^(cases/controls)(95% CI)(cases/controls)(95% CI)(cases/controls)(95% CI)TT/TCCCGGGA/AAAA/AGGGAge, years  < 5394/3015/27.997 (1.527–41.880)0.013986/29013/133.372 (1.507–7.546)0.003190/2949/93.267 (1.259–8.477)0.0150  ≥ 53193/2568/61.769 (0.604–5.181)0.2985168/24433/182.663 (1.451–4.886)0.0016189/26012/28.251 (1.826–37.289)0.0061Tumor size  < 2 cm165/5579/83.554 (1.323–9.548)0.0119156/53418/311.827 (0.984–3.393)0.0563160/55414/115.770 (2.495–13.348) < 0.0001  ≥ 2 cm122/5574/81.909 (0.556–6.549)0.304198/53428/314.562 (2.597–8.015) < 0.0001119/5547/113.239 (1.203–8.724)0.0201Family history Yes17/5571/84.417 (0.516–37.849)0.175316/5342/312.246 (0.491–10.264)0.296716/5542/115.987 (1.208–29.667)0.0284 No270/55712/82.794 (1.106–7.061)0.0298238/53444/312.952 (1.799–4.845) < 0.0001263/55419/114.375 (2.003–9.555)0.0002Pausimenia Post-menopause162/5576/81.584 (0.503–4.987)0.4315143/53425/312.362 (1.250–4.466)0.0082157/55411/118.180 (2.008–33.327)0.0034 Pre-menopause125/5577/84.609 (1.587–13.386)0.0050111/53421/313.718 (2.017–6.855) < 0.0001122/55410/113.380 (1.387–8.240)0.0074Clinical stage 124/5570/8 < 0.001 (< 0.001, > 999.999)0.986123/5341/310.689 (0.090–5.295)0.720022/5542/115.648 (1.139–28.004)0.0340 2174/5579/83.187 (1.183–8.588)0.0219159/53424/312.507 (1.408–4.462)0.0018170/55413/114.699 (2.002–11.027)0.0004 389/5574/82.843 (0.828–9.756)0.096872/53421/314.595 (2.489–8.484) < 0.000187/5546/114.117 (1.454–11.660)0.0077Pathological grade Low10/5340/31 < 0.001 (< 0.001, > 999.999)0.98638/5342/314.084 (0.826–20.195)0.084510/5540/11 < 0.001 (< 0.001, > 999.999)0.9844 High277/53413/313.052 (1.228–7.586)0.0163246/53444/312.850 (1.741–4.667) < 0.0001269/55421/114.621 (2.152–9.922) < 0.0001Distance metastasis Yes11/5570/8 < 0.001 (< 0.001, > 999.999)0.98955/5346/3117.478 (4.916–62.144) < 0.000111/5540/11 < 0.001 (< 0.001, > 999.999)0.9883 No276/55713/83.058 (1.233–7.584)0.0159249/53440/312.591 (1.569–4.278)0.0002268/55421/114.559 (2.128–9.766) < 0.0001Invasion Yes59/5573/83.161 (0.788–12.681)0.104444/53418/316.040 (3.083–11.834) < 0.000153/5549/1110.516 (3.929–28.143 < 0.0001 No228/55710/82.700 (1.037–7.026)0.0419210/53428/312.220 (1.288–3.826)0.0041226/55412/113.338 (1.422–7.835)0.0056Node infiltration Yes97/5574/82.761 (0.789–9.659)0.112174/53427/315.652 (3.154–10.128) < 0.000192/5549/116.128 (2.367–15.866)0.0002 No189/5579/82.837 (1.065–7.558)0.0370179/53419/311.734 (0.947–3.173)0.0745186/55412/113.886 (1.653–9.136)0.0019Molecular type Luminal A119/5574/81.837 (0.536–6.290)0.3331104/53419/312.839 (1.529–5.269)0.0009114/5549/114.851 (1.910–12.322)0.0009 Luminal B77/5575/84.268 (1.312–13.887)0.015971/53411/312.416 (1.143–5.106)0.020880/5542/111.795 (0.374–8.609)0.4644 Her2-positive44/5573/84.402 (1.118–17.339)0.034137/53410/314.509 (2.046–9.939)0.000243/5544/115.005 (1.514–16.537)0.0083 Basal-like47/5571/81.269 (0.153–10.536)0.825442/5346/312.253 (0.880–5.769)0.090342/5546/119.733 (3.240–29.243) < 0.0001ER expression Negative89/5574/82.843 (0.828–9.756)0.096877/53416/313.397 (1.764–6.542)0.000383/55410/116.871 (2.775–17.014) < 0.0001 Positive198/5579/82.803 (1.043–7.530)0.0410177/53430/312.642 (1.536–4.543)0.0004196/55411/113.550 (1.471–8.570)0.0048PR expression Negative91/5574/82.783 (0.812–9.545)0.103679/53416/313.313 (1.722–6.372)0.000385/55410/116.686 (2.704–16.531) < 0.0001 Positive196/5579/82.831 (1.053–7.609)0.0391175/53430/312.672 (1.553–4.597)0.0004194/55411/113.599 (1.490–8.695)0.0044Her2 expression Negative167/5575/81.717 (0.545–5.412)0.3562147/53425/312.662 (1.508–4.699)0.0007157/55415/115.924 (2.595–13.524) < 0.0001 Positive120/5578/84.402 (1.590–12.193)0.0043107/53421/313.176 (1.740–5.795)0.0002122/5546/112.922 (1.034–5.254)0.0430Ki67 expression  < 10%209/55710/82.923 (1.117–7.653)0.0289186/53433/312.863 (1.687–4.860) < 0.0001209/55410/113.045 (1.240–7.475)0.0151  ≥ 10%78/5573/82.450 (0.627–9.571)0.197668/53413/313.018 (1.494–6.097)0.002170/55411/119.013 (3.677–22.094) < 0.0001^a^ Adjusted for age

The rs16896059 polymorphism displayed a more substantial risk association among patients aged < 53 years (adjusted OR = 3.372, 95% CI 1.507–7.546, P = 0.0031), ≥ 53 years (adjusted OR = 2.663, 95% CI 1.451–4.886, P = 0.0016), with tumor size ≥ 2 cm (adjusted OR = 4.562, 95% CI 2.597–8.015, P < 0.0001), no family history (adjusted OR = 2.952, 95% CI 1.799–4.845, P < 0.0001), post-menopause (adjusted OR = 2.362, 95% CI 1.250–4.466, P = 0.0082) and pre-menopause (adjusted OR = 3.718, 95% CI 2.017–6.855, P < 0.0001), clinical stage 2 (adjusted OR = 2.507, 95% CI 1.408–4.462, P = 0.0018) and clinical stage 3 (adjusted OR = 4.595, 95% CI 2.489–8.484, P < 0.0001), high pathological grade (adjusted OR = 2.850, 95% CI 1.741–4.667, P < 0.0001), without distant metastasis (adjusted OR = 2.591, 95% CI 1.569–4.278, P = 0.0002) and with distant metastasis (adjusted OR = 17.478, 95% CI 4.916–62.144, P < 0.0001), with invasion (adjusted OR = 6.040, 95% CI 3.083–11834, P < 0.0001) and without (adjusted OR = 2.220, 95% CI 1.288–3.826, P = 0.0041), node infiltration (adjusted OR = 5.652, 95% CI 3.154–10.128, P < 0.0001), luminal A type (adjusted OR = 2.839, 95% CI 1.529–5.269, P = 0.0009), luminal B type (adjusted OR = 2.416, 95% CI 1.143–5.106, P = 0.0208), Her2 positive type (adjusted OR = 4.509, 95% CI 2.046–9.939, P = 0.0002), ER positively expressed (adjusted OR = 3.397, 95% CI 1.746–6.542, P = 0.0003) and negatively expressed (adjusted OR = 2.642, 95% CI 1.536–4.543, P = 0.0004), PR positively expressed (adjusted OR = 3.313, 95% CI 1.722–6.372, P = 0.0003) and negatively expressed (adjusted OR = 2.672, 95% CI 1.553–4.597, P = 0.0004), Her2 positively expressed (adjusted OR = 2.662, 95% CI 1.508–4.699, P = 0.0007) and negatively expressed (adjusted OR = 3.176, 95% CI 1.740–5.795, P = 0.0002), and Ki67 expressed cells ≥ 10% (adjusted OR = 3.018, 95% CI 1.494–6.097, P = 0.0021) and < 10% (adjusted OR = 2.863, 95% CI 1.687–4.860, P < 0.0001).

The rs2449512 polymorphism exhibited a more substantial risk association among patients aged older than 53 years (adjusted OR = 8.251, 95% CI 1.826–37.289, P = 0.0061), younger than 53 years (adjusted OR = 3.267, 95% CI 1.259–8.477, P = 0.0150), with tumor size smaller than 2 cm (adjusted OR = 5.770, 95% CI 2.495–13.384, P < 0.0001) and larger than 2 cm (adjusted OR = 3.239, 95% CI 1.203–8.724, P = 0.0201), possessing a family history (adjusted OR = 5.987, 95% CI 1.208–29.667, P = 0.0284) or no family history (adjusted OR = 4.375, 95% CI 2.003–9.555, P = 0.0002), being in post-menopause (adjusted OR = 8.180, 95% CI 2.008–33.327, P = 0.0034) or pre-menopause (adjusted OR = 3.380, 95% CI 1.387–8.240, P = 0.0074), presenting clinical stage 1 (adjusted OR = 5.648, 95% CI 1.139–28.004, P = 0.0340), clinical stage 2 (adjusted OR = 4.699, 95% CI 2.002–11.027, P = 0.0004) or clinical stage 3 (adjusted OR = 4.117, 95% CI 1.454–11.660, P = 0.0077), characterized by a high pathological grade (adjusted OR = 4.621, 95% CI 2.152–9.922, P < 0.0001), without distant metastasis (adjusted OR = 4.559, 95% CI 2.218–9.766, P < 0.0001), invasion (adjusted OR = 10.516, 95% CI 3.929–28.143, P < 0.0001) or without (adjusted OR = 3.338, 95% CI 1.422–7.835, P = 0.0056), node infiltration (adjusted OR = 6.128, 95% CI 2.367–15.866, P = 0.0002) or without (adjusted OR = 3.886, 95% CI 1.653–9.136, P = 0.0019), and luminal A type (adjusted OR = 4.851, 95% CI 1.910–12.322, P = 0.0009), Her2-positive type (adjusted OR = 5.005, 95% CI 1.514–16.537, P = 0.0083) and basal-like type (adjusted OR = 9.733, 95% CI 3.240–29.243, P < 0.0001), ER negatively expressed (adjusted OR = 6.871, 95% CI 2.775–17.014, P < 0.0001) and positively expressed (adjusted OR = 3.550, 95% CI 1.471–8.570, P = 0.0048), PR negatively expressed (adjusted OR = 6.686, 95% CI 2.704–16.531, P < 0.0001) and positively expressed (adjusted OR = 3.599, 95% CI 1.490–8.695, P = 0.0044), Her2 negatively expressed (adjusted OR = 5.924, 95% CI 2.595–13.524, P < 0.0001) and positively expressed (adjusted OR = 2.922, 95% CI 1.034–5.254, P = 0.0430), and Ki67 expressed cells < 10% (adjusted OR = 3.045, 95% CI 1.240–7.475, P = 0.0151) and ≥ 10% (adjusted OR = 9.013, 95% CI 3.677–22.094, P < 0.0001).

Haplotype analysis of three MTDH gene SNPs correlated with IDC susceptibility

We further investigated the potential association between the haplotypes of the three MTDH gene SNPs and the risk of IDC. As illustrated in Table 4, the wildtype allele TGA was designated as the reference group. Compared to the reference haplotype TGA, the following haplotypes were found to be significantly associated with an increased risk of IDC: TAA (adjusted OR = 6.983, 95% CI 3.788–12.874, P < 0.001), TAG (adjusted OR = 2.392, 95% CI 1.007–5.682, P = 0.048) and TGG (adjusted OR = 1.584, 95% CI 1.150–2.181, P = 0.005). Table 4. Association between inferred haplotypes of the MTDH gene and IDC riskHaplotypes^a^Cases (n = 300)Controls (n = 565)Crude OR (95% CI)PAdjusted OR^b^ (95% CI)PNo.%No.%TGA404 (23.35)883 (51.04)1.0001.000CGA45 (2.60)98 (5.66)1.004 (0.692–1.456)0.9850.986 (0.679–1.431)0.940TAA45 (2.60)14 (0.81)7.025(3.812–12.946) < 0.0016.983 (3.788–12.874) < 0.001CAA6 (0.35)8(0.46)1.639 (0.565–4.755)0.3631.585 (0.545–4.607)0.398CAG1 (0.06)0 > 999.999 (< 0.001, > 999.999)0.979 > 999.999 (< 0.001, > 999.999)0.978TAG11 (0.64)10 (0.58)2.404 (1.013–5.707)0.0472.392 (1.007–5.682)0.048TGG76 (4.39)103 (5.95)1.613 (1.172–2.218)0.0031.584 (1.150–2.181)0.005CGG12 (0.69)14 (0.81)1.873 (0.859–4.087)0.1151.810 (0.828–3.956)0.137^a^The haplotypes order was rs1311, rs16896059, and rs2449512^b^Obtained in logistic regression models with adjustment for age

Discussion

In the present case–control study, involving 300 IDC cases and 565 healthy controls from Eastern Chinese populations, we investigated the potential association between MTDH gene polymorphisms and the risk of IDC. Our findings provide evidence that three polymorphisms, namely rs1311 T > C, rs16896059 G > A, and rs2449512 A > G, were associated with an increased susceptibility to IDC. Notably, our study is the first to establish the association between MTDH polymorphisms and IDC susceptibility.

MTDH plays a crucial role in various stages of carcinogenesis and serves as an oncogene in multiple cancers. In breast cancer, MTDH acts as a mediator for numerous non-coding RNAs. For instance, the long non-coding RNA (lncRNA) FAM83H-AS1 facilitates the progression of triple-negative breast cancer through binding miR-136-5p to increase MTDH expression [27]. Knockdown of lncRNA TP73-AS1 inhibits in vitro breast cancer cell carcinogenesis by targeting miRNA-125a-3p to suppress MTDH levels [28]. LINC00707 directly targets MTDH to inhibit breast cancer by sponging miR-876 [29]. Furthermore, anti-cancer agents can target MTDH to suppress breast cancer. Lobaplatin inhibits cell proliferation and induces apoptosis by downregulating MTDH in breast cancer [30]. The tumor suppressor FBXW7 also hinders breast cancer proliferation and promotes apoptosis by degrading MTDH [31]. Overexpression of MTDH is associated with doxorubicin sensitivity of breast cancer [32]. Furthermore, the expression of MTDH is prognostically linked to diverse molecular subtypes among patients with breast cancer following therapeutic intervention. Li Y et al., employed Affymetrix microarrays to identify genes that are differentially expressed between estrogen-treated parental cells and those deficient in MTDH. Subsequently, they determined that MTDH and ERα interact within the nucleus under estrogenic treatment to regulate gene expression [33]. Chu PY, et al. demonstrated that MTDH serves as an independent prognosticator of inferior outcomes in patients with ER-negative or PR-negative breast cancer [14]. Elevated MTDH expression predicts a better prognosis for HER-2 positive breast cancer patients following combined therapy of neoadjuvant chemotherapy and trastuzumab [15]. Despite numerous studies exist regarding the function of MTDH in breast cancer, the association between MTDH polymorphisms and breast cancer risk remains unreported. However, one study did reveal a negative association between the MTDH (− 470G > A) polymorphism and ovarian cancer susceptibility [21].

In the current study, genotyping was performed on three SNPs of MTDH, rs1311, rs16896059, and rs2449512. The results revealed a significant association between the genotypes of rs1311 T > C, rs16896059 G > A, and rs2449512 A > G with an increased risk of IDC. Our findings suggested that the presence of the rs1311 C allele, rs16896059 G allele, and rs2449512 G alleles exacerbated the IDC risk in Eastern Chinese women. Furthermore, all three polymorphisms were found to contribute to elevate IDC risk in women aged 53 years or younger, without a family history, with pre-menopause status, clinical stage 2, without distant metastasis or invasion. These associations were observed specifically in patients with Her2-positive type, ER positive, PR positive, and Ki67 cells greater than 10%. However, the effects of rs16896059 and rs2449512 on IDC risk were more prominent in patients with tumor size larger than 2 cm, post-menopause, clinical stage 3, low pathological grade, invasion, node infiltration, ER negative, PR negative, Her2 negative, and Ki67 cells less than 10%. Additionally, these results indicated that the genotype rs1311 A > G could serve as a reference for IDC subtyping and therapeutic decision-making. Based on previous research [34], haplotypic association studies involving multiple SNPs have been shown to significantly enhance gene mapping power compared to single SNP studies when it comes to identifying disease-causing genes. Consequently, we investigated whether haplotypes composed of MTDH gene polymorphisms rs1311, rs16896059, and rs2449512 were associated with IDC risk. Our findings revealed that these variants could interact with each other to influence the risk of IDC.

The effect and function of these three selected SNPs in MTDH gene has not been reported. Since rs1311 and rs2449512 are located in the 3ʹUTR of MTDH gene, they were predicted to affect miRNAs binding. Non-coding RNAs regulate the expression and function of MTDH by affecting their binding to miRNAs in breast cancer. For instance, circHIPK3 in breast cancer-derived exosomes promotes angiogenesis in the tumor microenvironment through elevating the expression of MTDH by sponging miR-124-3p [35]. OTUD6B-AS1 facilitates paclitaxel resistance by sponging miR-26a-5p to upregulating MTDH, then promoting autophagy and genome instability [36]. MiR-9-3p enhances the drug resistance of gemcitabine by directly targeting MTDH [13]. A specific H3K79 methyltransferase DOT1L increase the MTDH expression by increasing H3K79me3 levels on its promoter to promoting angiogenesis in triple-negative breast cancer [37]. As rs16896059 is located in the promoter of MTDH gene, its polymorphism might regulate transcriptional activity. In the future study, we need to perform experiments to verify the effects of these three SNP polymorphisms on the expression and function of MTDH.

There are still several limitations in this study. Firstly, the sample size was inadequate. Secondly, being a retrospective study, it inevitably resulted in information bias and selection bias. To mitigate these biases, we employed frequency matching of cases and controls based on age and BMI. Thirdly, the samples were recruited from a single center, potentially introducing unavoidable selection bias. This study primarily focused on the analysis of genetic variations associated with IDC susceptibility. However, crucial information such as environmental factors, gene mutations, breastfeeding, and lifestyle was unavailable for analysis. Lastly, the relationship between MTDH gene polymorphisms and the prognosis of IDC was not examined in the current study.

In summary, our findings demonstrated a significant association between polymorphisms rs1311 T > C, rs16896059 G > A, and rs2449512 A > G in the MTDH gene and an increased risk of IDC in Eastern Chinese women. Further investigation was warranted to elucidate the biological function of these MTDH gene risk SNPs in the etiology of IDC. Our results indicated that polymorphisms in the MTDH gene were linked to a heightened susceptibility to IDC. These findings suggested that MTDH gene polymorphisms had potential as biomarkers for assessing IDC susceptibility.